At the dawn of World War II, Konrad Lorenz (1941), an Austrian biologist, published an article that proposed a radical new perspective on traditional epistemological questions by situating them in a Darwinian context. In this article, Lorenz attempts to provide an evolutionary account of the working of our cognitive mechanisms and points out the consequences that this holds for our epistemic endeavours. While similar approaches have been hinted at by Herbert Spencer (1855), Charles Darwin (1871) himself and famous pragmatists such as William James (1890) and John Dewey (1910), Lorenz was the first to elaborate this view and to offer empirical evidence in support of it.
Despite its importance and implications, however, it remained largely unknown for over three decades. The world was at war, Lorenz was Austrian, and to make matters worse he joined the Nazi party and published pseudo-scientific writings supporting Nazi ideology. However, more than three decades later, Lorenz’s groundbreaking work in the field of biology was recognised, and he was awarded a Nobel Prize, together with his colleagues Nikolaas Tinbergen and Karl von Frish. He officially apologised for his previous political opinions and was embraced by the world as one of the founders of a new science: ‘ethology’ - the study of animal cognition and behaviour.
In this section, I will outline what I refer to as ‘the Lorenzian argument’, at the core of evolutionary epistemology10, stating that:
10 When referring to evolutionary epistemology, I aim at those theories reasoning about human knowledge from
the premise that the cognitive abilities underlying it are the outcome of biological evolution. As pointed out (cf. General introduction), a distinct program also runs by the name of evolutionary epistemology, drawing an analogy between the evolution of ideas or theories and biological evolution by means of blind variation and selective retention. These theories, developed by thinkers as Popper (1972) and Campbell (1974), are not our concern here.
Since our perceptual and cognitive apparatus, providing us with a representation of the world, are the result of evolution by natural selection, the structures of this representation must – at least approximately – match the structures of the world itself, because being endowed with an accurate representation of the world increases one’s biological fitness and therefore the perceptual and cognitive mechanisms allowing for this accurate representation must have been selected.
2.1 Ontogenetic a priori as a phylogenetic a posteriori 2.1.1 Adaptations as representations
Lorenz (1941) starts his argument by stating that all features of organisms ‘mirror’ nature. Indeed, he claims, the process of natural selection has adapted these features to the environment in which the organism lives. The eye reflects the laws of optics in its very structure, the shape of a fish reflects the laws of hydrodynamics, and the anatomy of a bird’s wings those of aerodynamics. All characteristics of an organism, shaped by the process of natural selection, reflect the elements of the environment to which they are adapted.
In this regard, Peter Munz (1993) views organisms as ‘embodied theories’. The organism, he claims, incarnates theories about the environment - it reflects ‘knowledge’ about the environment. According to Munz, however, this does not mean that the organism mirrors or even describes those parts of the world to which it is adapted, but merely, that some properties of the organism disclose information about certain properties of the environment. The hydrodynamic shape of fish does not describe water, let alone mirror it, but tells us something about the properties of water in relation to locomotion. This representation, of course, is neither verbal nor conscious. It is expressed through an organism’s anatomical structure and behavioural programming.
In this sense, it is embodied, not disembodied – meaning that it cannot be separated from the organism expressing it. When the kingfisher dives into the water to catch a fish, it does so by correcting the angle of refraction. However, it does not know Snell’s law – calculating the angle of the refraction of light when it passes from air to water – but, that knowledge is,
unconsciously and in a non-verbal, non-mathematical way, present in its genetic constitution, triggering a particular action-scheme when it spots a fish from above (154-158). This is what is meant by ‘incarnated knowledge’ or ‘embodied theory’. Adaptations, in this perspective, are representations of aspects of the world.
2.1.2 A non-arbitrary a priori
Lorenz (1941) argues that human perceptual and cognitive structures are no exception. Our mental a priori, structuring data into (conscious) knowledge, has evolved through natural selection in the same way as every other characteristic or ability of any organism. This implies, according to Lorenz, that they are not arbitrary. Just as any other feature of an organism reflects or represents the structures of the elements of the environment to which they are adapted, so do our perceptual and cognitive abilities.
Just as the hoof of the horses is adapted to the ground of the steppe which it copes with, so our central nervous apparatus for organizing our image of the world is adapted to the real world with which man has to cope. Just like any organ, this apparatus has attained its expedient species-preserving form through this coping of real with the real during a species history many eons long (Lorenz, 1941:124).
In other words, our perceptual and cognitive systems are shaped by natural selection to cope with the external world. Therefore, he argues, it must provide us with an accurate representation of reality, allowing us to make true inferences about what really is the case. If, indeed, our mental abilities did not represent reality in a truthful way, Lorenz argues, they could never have evolved. Organisms endowed with perceptual and cognitive structures that did not lead to a truthful representation of reality would have less than average chances for survival and reproduction. Therefore the genes responsible for these inaccurate representations of reality would soon be weeded out of the gene-pool. Natural selection shapes every organism to ‘fit’ their environment - similarly it has shaped our mental a priori to ‘fit’ the world around us, i.e. to yield a truthful representation of objective reality.
In this context, Lorenz casts the Kantian a priori – our innate perceptual and analytical structuring of external data – in an evolutionary light. Lorenz agrees with Kant that we are endowed with a mental a priori - we do not merely absorb reality as it is, but actively structure it, both through perception and understanding. He denies, however, that these perceptual and analytical structures are necessary and universal. According to Lorenz, they are the product of evolution by natural selection and are therefore ‘tuned’ to the external constraints of the world. This entails that they are no longer necessary, nor arbitrary – as Kant considered them – but become particular and reliable. Indeed, whereas Kant considered the structures through which we mould reality as universal – every rational or thinking being would, in his opinion, perceive and understand reality through those structures – but arbitrary – there was no way of conceiving why we structure reality the way we do - Lorenz explains this a priori as the result of natural selection. In this regard, it is no longer necessary or universal, since it is an adaptation of a particular organism, but it also loses its arbitrary character, for it must be shaped to provide us with an accurate representation of the world.
2.1.3 The experience of the lineage
Therefore, what is a priori – prior to experience – from an ontogenetic level (the level of the organism), becomes a posteriori – posterior to experience – on a phylogenetic level (the level of the species) (Wuketits, 1990). We might look at reality through spectacles that structure it and therefore not see the world as it is in itself, but evolution by natural selection entails that the world itself shaped these spectacles, and therefore ensures that the picture of reality they provide us with is a truthful one. The particular way in which we perceive and understand reality is the result of the experience gathered by our species through evolution. Accurate ways of representing reality are kept at the expense of inaccurate ones. As Lorenz phrases it:
The ‘spectacles’ of our modes of thought and perception, such as causality, substance, quality, time and place, are functions of a neuro-sensory organization that has evolved in the service of survival. When we look through these ‘spectacles’, therefore, we do not see, as transcendental idealists assume, some unpredictable distortion of reality which does not correspond in the least with things as they really are, and therefore cannot be regarded as an image of the outer world (Lorenz, 1977:7).
The Lorenzian argument, therefore, rejects both empiricism and transcendental idealism. As Buskes (1998) explains, the empiricist’s tenet, claiming that all knowledge is derived from experience, is false, because humans – like any other organism for that matter – are endowed with innate, a priori knowledge. Kant’s transcendental idealism, on the other hand, which argues that the world in itself is necessarily beyond our ken since it is moulded by the subject, is – according to the Lorenzian epistemologist – equally mistaken, because our perceptual and cognitive apparatus conform to a pre-structured world by the process of natural selection (43). Indeed, according to Hoffmeyer (1993), with every generation, a meeting between the species and its environment occurs – each time with a different outcome – as every generation passes on a different set of genetic material to the next generation. The genes that provide an organism with the best ways of coping with the environment will be more present in the next generation than in the previous one, because the organisms endowed with these genes are more successful in reproducing. Every single generation, however, is faced with a unique set of conditions, since environments are constantly changing. This means, from the point of view of the lineage or the gene-pool of the species, that it constantly adapts in response to the experience it gathers from the confrontation of every generation with the environment (21- 22).
This ongoing experience of the lineage through consecutive generations, it is argued, ensures a correspondence between cognition and nature. While we are endowed with an a priori (an innate way of perceiving and understanding the world; in other words, the ‘spectacles’ through which we view the world) that we cannot transcend on the ontogenetic level, the process of evolution by natural selection entails that this a priori is the product of a posteriori knowledge (i.e. knowledge gained from experience) on the phylogenetic level. The lineage, in other words, acquires knowledge of the ‘world in itself’ through experience, and this knowledge is reflected in the working of our perceptual and cognitive abilities. The representation of the world they provide us with must, therefore, correspond to the world itself.
In this regard, the view of the world that emerges from our perceptual and cognitive abilities can be seen, according to Vollmer (1984), as hypotheses about the structures of the real world. These hypotheses are unconscious and uncritical, and therefore utterly incorrigible. However, through the process of evolution by natural selection, (heavily) mistaken hypotheses soon die out and are replaced by better trials and better conjectures, enhancing the chances of survival and reproduction. Indeed, if these ‘a priori conjectures’ about the world are relevant for survival and reproduction, natural selection will gradually improve them, as it does for any other characteristic of the organism. This process of trial and error; mutation and selection, must eventually lead to a far-reaching similarity between an organism’s representation of reality and the objective structures of reality (75). So, at least, goes the argumentation of many evolutionary epistemologists (see also Wuketits, 1990:88).
2.1.4 The enigmatic congruence between mind and the world
This view of natural selection ensuring the correspondence between our mental a priori and the external world, offers an explanation to a philosophical conundrum that has baffled philosophers for ages and that has led to a wide variety of theories. How is it possible that our mental structures apply to the external world? How, in other words, can external events and properties be predicted on the basis of a priori, mental models? Why is – as Galileo (1623) phrased it – the great book of nature written in the language of mathematics?
Plato (380 B.C.) thought this ability originated in the pre-existence of the soul, which contemplated the Ideas – the essence of reality – freely, before being cast in an earthly body. Descartes (1641) ‘proved’ the existence of a perfect God – and therefore one that would not deceive us – to explain why our innate ideas enable us to understand reality. Leibniz (1704) invoked a ‘pre-established harmony’ between the subject and the objective world as given by the hand of God. Kant’s (1781) transcendental philosophy turns the tables on these traditional approaches, and holds that our cognitive structuring does not match the world, but rather that it is the experienced world – through the categories of perception and understanding – that matches our cognitive structuring. Our mind – in other words – is not shaped to unveil the objective structures of reality, but our subjective experience of reality is shaped in a way that allows us to acquire knowledge of it. Lorenzian evolutionary epistemology, on the other hand,
explains this congruence between our cognitive structures and external reality through the process of adaptation. Our mental a priori, it claims, is the product of a process shaping it in accordance with the external world.
2.2 Hypothetical realism
The claim that the representation of the world that we acquire through our ‘a priori veil’ – i.e. through our perceptual and cognitive apparatus – matches the structures of the real world, can therefore, according to the Lorenzian argument, not be doubted. In the meantime – as Lorenz (1941, 1973) was fully aware – natural selection does not ensure an autonomous and absolute validity of man’s grasp of reality. They do not have a priori validity; they are not necessarily true, nor infallible, but merely provide us with decent approximations of what really is the case. In other words, they provide us with a truthful ‘basic’ representation of the world, but not one that mirrors the world in all its inherent complexity.
This approximate character of cognition is elegantly illustrated by Lorenz in the following comparison:
[Just as] the ‘dots’ produced by the coarse ‘screens’ used in the reproductions of photographs in our daily papers are satisfactory representations when looked at superficially, but cannot stand closer inspection with a magnifying glass, [so], too, the reproductions of the world by our forms of intuition and categories break down as soon as they are required to give a somewhat closer representation of their objects (Lorenz, 1941:128).
Therefore, it is argued, evolutionary epistemology leads to ‘hypothetical realism’ (Vollmer, 1984:78). It concludes, from the fact that our cognitive system is shaped by natural selection, that there is a far-reaching agreement – albeit an incomplete and imperfect approximation – between the objective structures of the world and our mental reconstruction of those structures through perception and cognition.
Vollmer (1984) points out the impossibility of a perfect agreement. Such an ideal adaptation, he claims, is not necessary for survival, and could only occur at a very high cost. Evolutionary change does not come for free. Homo sapiens’ increased brain power necessitated an equally increased amount of nutrients to keep the organism going. Therefore, there is always a trade- off between advantages procured by a new ability or physical trait and the cost of this adaptation. Furthermore, he claims, a perfect fit would mean a very rigid ‘reflection’ of nature, leaving no chance of survival in the case of environmental changes (78).
This hypothetical realism forms the backbone of Lorenzian evolutionary epistemology. Admitting that our representation of the world might never match the world itself perfectly, it argues that it must nevertheless be a decent approximation of it (see Lorenz 1941, 1971; Vollmer 1975, 1984; Riedl 1984b; Oeser 1987; Wuketits 1990). This train of thought is at the core of Quine’s famous quote with respect to the infamous problem of induction.
If people’s innate spacing of qualities is a gene-linked trait, then the spacing that has made for the most successful inductions will have tended to predominate through natural selection. Creatures inveterately wrong in their inductions have a pathetic but praiseworthy tendency to die before reproducing their kind (Quine, 1969:126).
At first glance, within the framework of evolution by natural selection, this claim seems very plausible. How, indeed, could a more accurate representation of the world - a more correct way of inducing - not profit an organism in terms of survival and reproduction? However, as I will argue, this view commits the fallacy of deducing the truthfulness of our representation of the world from its mere functionality with regards to biological fitness. In order to show this, I will first point out that natural selection, only shaping perceptual and cognitive abilities to produce desired behaviour (cf. previous chapter), cannot be expected to yield true beliefs about the world automatically. In doing so, I will give an overview of the possible relations between beliefs and behaviour, based on Plantinga’s (1993) argument against evolutionary naturalism. Moreover, I will argue, the assumption that true inferences have more survival and reproductive value than false ones is not as self-evident as it appears. Furthermore, a closer look at variation as the motor of adaptation, and the importance of both the particular elements of the environment and previously acquired traits in the selection of new ones,
points to the constitution of an imperfect and very particular, species-specific way of representing reality.